One of the Rational insiders who reviewed my article before it was published in last month's issue¹ took me to task for not sufficiently condemning "politics" as one of the root causes of difficulty in the software construction process. This revealed to me a misunderstanding about the role of politics, not just in software development, but also in the management of technical organizations in general. The purpose of this piece is to explore the many dimensions of politics in technical organizations.

Context

Before launching headlong into the meat of the matter, I need to say a few things about context. Any time we talk about human behavior, such as "politics" and political process, we enter into the minefield of cultural differences. Simply put, different behaviors are more or less acceptable in different cultural contexts. Behavior that could be considered "borderline" in one culture could be considered so offensive in another that it would never happen. When we get into "good politics" and "bad politics" below, let's remember that the norm I am using here is a North American standard, and, even at that, susceptible to many local gradations. Some of the behaviors described might be more or less acceptable in European or Asian contexts. The part of the discussion that I believe transcends national and cultural boundaries is the difference in the way technical people (and the organizations they belong to) view the political process. Here I have observed some common tendencies that are worth noting, independent of geography and culture.

Highly competent technical people often have a distaste for politics and politicians. Although I'm no psychologist, I believe this may stem from their notion that technical matters are "precise" (black and white) whereas politics is "messy" (lots of shades of gray.) By training and education, scientists and engineers have a very analytic, problem-solving approach, which involves invoking "first principles" and applying data in the course of various inductive and deductive processes. Often, this leads to neat, "closed form" solutions. And because many technical managers come from the ranks of scientists and engineers, they bring along with them not only this very strong toolbox, but also some of the biases that come along with it.

Politics involves a great deal of ambiguity, which makes many technical people nervous. It forces them to play on "someone else's court," as it were, a place where they feel they are at a disadvantage because their technical prowess doesn't count for much.

While this is perhaps an oversimplification, it is fairly safe to say that technical people would prefer issues to be resolved on technical merits alone, or perhaps on some mix of technical and business objectives. What they react to, often violently, is the manipulation they see in various discussions and negotiations -- a manipulation they associate with the darker, uglier side of human behavior -- often characterized by self-aggrandizement and personal agendas. Technical people have a tendency to broad-brush this behavior as "politics," and in so generalizing it, give it a very negative connotation.

On the other hand, regardless of its nature, just wishing "politics" away will not make it go away. There is "the way we would like things to be," and the way things are. Ironically, some people view bridging this gap as itself a political process.

My point is that so long as you have human beings in the loop, you are going to have politics. Human beings tend not to all think alike; in order to resolve differences of opinion, a political process is unavoidable. So, rather than condemn it, it is better to understand politics as an effective means of dealing with the inevitable need to resolve differences of opinion.

Back to top

Definition

Let us note that the word "politics" refers to a category of human activity like "carpentry," "theater," or "surgery." What we really need to define is the term "political process." A political process is one in which you get a person or group of people to do what you want them to do. This is actually a "shorthand" for a much more complicated set of ideas.

A political process is one in which two or more people adopt a single course of action. Typically, each individual involved in a political process maintains an agenda that is different from the group's collective agenda, or from any other individual's personal agenda. It is through the political process that an individual's agenda becomes publicized (if a free press is at work), negotiated and voted on (if a democracy), or implemented without further discussion (if the individual is the king). In every case, the individual engaged in a political process wants the group to adopt his or her agenda.²

By the way, getting other people to do what you want them to do is also the job of leaders and managers. But you probably have already made that observation yourself.

Note that "what" you are trying to convince them to do can be either "good" or "bad"; that is irrelevant to the discussion. Of course, from your point of view, you are probably trying to get them to "do the right thing," but this certainly depends on your perspective. Even when purely motivated, your objectives may be viewed as "wrong," "bad," "misguided," or even "evil" by some other third party.

Back to top

Three Scenarios

If you are in the category of people who still find politics not their cup of tea, let me give you another way to look at it. There are, in fact, only three distinct cases to consider.

1. You Are the King

You might think that politics is irrelevant in this case. The simplistic view might be that even if you are a benevolent despot, you pretty much get to do what you want, and your vassals obey. If politicians crop up, you suppress them; dissident serfs get thrown in the oubliette³, so you don't have to deal in politics. In a business organization, the equivalent behavior is to fire any employee who shows "political" tendencies.

But this is an oversimplification. Even kings need to build some support amongst their circle of advisors in order to stay in power. There is a good reason today why we don't have very many absolute, powerful kings left. But, if you are the king and want complete control over your realm, you can keep politics at bay for a while. In today's world, it is just not a really good, viable, long-term strategy.

2. You Are Not the King

Now you have two choices: one, convince the king to do what you want to do, or, two, band together with other vassals to either reason with the king or overthrow him. I will point out that both alternatives imply a political process. If you are unclear on this concept, go back to our earlier definition of a political process above.

3. There Is No King

This often occurs in organizations where there is no strong leadership. You might even call this nascent democracy. In this instance, people have to get together and decide what to do, as there is no king telling them what to do. People will tend to have differing opinions, so we once again get to decision making by consensus building (also a political process).

Whether or not you consider technical organizations "democracies" (and we all know managers who strongly believe they are not!), it is important to understand that we exist in a culture that encourages participation of all its members. In fact, the strongest technical organizations are the ones that engage the talents of all, in a free marketplace of ideas. Ideally, one wants to create a high-trust environment⁴, that is, an environment where there is a high degree of trust among employees at every level, and where intellectual honesty reigns and politics is minimized. But we've just shown that politics is inevitable in all organizations, and technical organizations are just another form of organization. How do we deal with this apparent paradox?

Back to top

Politics Is Inevitable, But

In almost every organization, there is going to be some degree of politics. In the House of Representatives and the Senate, political context and the bartering that accompanies it are fundamental. But as we have pointed out above, when it comes to technical organizations, most would like to minimize the political context. Why?

The answer is simple. You cannot solve technical problems by political compromise. If there are two sides to an issue, one of which is technically feasible and one of which is technically ridiculous, no political compromise will work. Imagine an absurd example: The "solidistas" want dams to be solid, while the "holeistas" want dams to look like Swiss cheese. The political compromise of having only one or two holes in the dam would make no sense whatsoever.

Most people quickly understand the issues implicit in the last paragraph, but not always. I have seen, on occasion, otherwise intelligent people falling into the trap of trying to mediate between competent and incompetent people, giving equal weight to their ideas. This never works, annoys both parties, and tends to really discourage the competent people. Another way of saying this is that "Everyone is entitled to an opinion," but technical people know that not all opinions are equally well qualified.

Another, more subtle variant of the political pitfall for technical organizations is bartering. In many political processes, two sides get to trade on issues: You concede on this one and I'll concede on that one. We each "win one," and we each "lose one." This is the most frequent way of avoiding deadlocks.

Well, in technical organizations this won't work in many contexts. If there are two dams to be constructed, to reuse the example above, then again it makes no sense to build dam "A" according to the "solidista" philosophy and dam "B" according to the "holeista" philosophy. One will hold water and the other won't. Period.

Nor does it work to give the wrong-headed side free rein in some other area. If the political compromise to achieve solid dams is to let the incompetent "holeistas" go off and design another class of structure -- say fortresses -- then we have once again made a really bad decision through the use of "politics."

Of course, the choices in real life are not always so clear. If it were always a choice between the "solidistas" and the "holeistas," our lives would be easier. Often the two (or more) alternatives are not so obviously differentiable, and the technical issues may be very muddy indeed. In such cases, it is not always true that both sides are willing to admit that the other side has some merit. Often, technical criteria become submerged and the discussion degenerates into politics. This is generally bad.

Back to top

When Things Get Political

So what is the legitimate place for politics in technical organizations?

Let's agree that purely technical issues need to be resolved technically. Once an issue moves out of the purely technical realm, then we admit politics. Examples of legitimate political issues include decisions on marketability, customer satisfaction, business impact, and so on. Since objective, technical criteria are harder to pin down in these areas, debate, including political debate, is necessary.

However, there are "good politics" and "bad politics." Political scientists would, of course, disagree: For them, all methods are more or less efficacious in their own ways, and they are loath to make value judgments. They might consider my categorizations naîµ¥. So be it. Let's explore "good politics" and "bad politics" a little bit further.

Good Politics

The following techniques can be considered valid, legitimate parts of a political process:

1. Education
2. Persuasion
3. Consensus building
4. Fact-finding
5. Intellectually honest discussions
6. Identification of common interests
7. Exposure of hidden or subtle facts
8. Seeking compromise
9. Reasoning together

Plato could get behind this kind of politics. In fact, it has been observed that in organizations that do a good job of consensus building, decisions often are implemented more smoothly, because the consensus-building process causes diverse factions to come together and exercise reason, and to participate in making a joint decision. This process in effect "pre-sells" the idea, and the eventual solution has more buy-in from the participants than if there had not been consensus building.

By the way, I'm proud to observe that one of the hallmarks of the early Rational culture was an emphasis on consensus building to arrive at good decisions. A little later we needed to introduce "time bounded consensus seeking," wherein if consensus was not achieved by a certain time, a decision was made through a somewhat less democratic, more informal process by the relevant manager, in the interest of moving on. Even in this case, the period during which consensus was sought did tend to give a good airing of all the alternative points of view.

Neutral Politics

The following behaviors are in the gray zone. Some people admit them as part of a legitimate political process, while others abhor them. I make no value judgment here other than to put them in the "neutral" category⁵:

1. Cajoling
2. Ridiculing
3. Lobbying
4. Delaying
5. Defocusing issues
6. Positioning
7. Not telling all the truth all the time
8. You do this for me and I'll do that for you

Number eight is basically yielding on one point to gain another, what we called "bartering" above. One man's "horse trade" is another man's "compromise." This is why I place number eight in the gray zone.

Bad Politics

Without belaboring the neutral zone too long, we pass to those aspects of the political process that most people find unpleasant and "over the line":

1. Lying or deliberately misleading
2. Bribing
3. Intimidating, threatening, bullying
4. Undermining, conspiring, plotting
5. Personal attacks, abusive behavior
6. Filibustering
7. Hidden agendas
8. Committing to do something you have no intention of doing
9. Committing to not do something you have every intention of doing
10. Appeal to authority to subvert the process
11. "The end justifies the means," or "All's fair in love and war."

Sometimes this brand of politics is labeled "Machiavellian." This does a great disservice to Machiavelli, who had a lot more to say than "The ends justify the means."⁶

What happens when you get to bad politics is that you now are dealing with people who don't view politics as an adjunct to getting their jobs done. You are dealing with people for whom the political process itself is the primary preoccupation. Winning, by political or other means, is more important to these people than getting the job done. They have their priorities reversed, and, as such, are damaging to the organization. Certainly when you get to number eleven, which basically says, "There are no rules in a knife fight," you are beyond the pale.

The Engineering Mapping

The real problem with the gray zone is that many technical people have a bias toward the very high end of the integrity spectrum. They may put gray zone behaviors in their category of "bad politics." For example, not telling the whole truth all the time is technically "lying," yet some people admit a spectrum here, while others view it as black or white. It is difficult to have a "high trust" environment if there is a lot of gray zone behavior.

So don't be surprised, when you talk about "politics" with engineers and technical managers, if they lump "neutral (gray) zone" and "bad" politics under the general, derogatory heading of "politics" and characterize the "good" politics category as "leadership" or "good management." This turns things into a black and white world where all "politics" is evil. As this was my starting point, I now want to make it clear that we can sort things out, and what remains is really a linguistic mapping issue. The following table may prove helpful:

One could say that the entire thrust of this article is reconciling the "mainstream usage" mapping to the "engineering" mapping. These are two different ways of looking at the world, but without understanding the linguistic overloading we can get into a lot of trouble.

Especially when those engineers think we are "defending politics."⁷

High Trust Environments

We've mentioned "high trust environment" in the context of an organization's culture. Perhaps we should be explicit about what it is, and why it is so desirable.

In a high trust environment, the following is taken as the norm:

• We can trust each other to tell the truth;
• We can depend on each other to do whatever it takes;
• We can depend on each other to put the organization's objectives ahead of our own personal or group objectives;
• We can assume intellectual honesty in all discussions;
• We don't take commitments lightly -- we view missed commitments as a violation of trust. This implies both volition and competency.

The thing that is so wonderful about high trust environments is that they are extremely efficient. When you have trust, you have less need for verification. This means you can get more done with smaller teams. It is the equivalent of reducing friction in a machine: More of the energy goes to producing work (good stuff) and less to producing heat (bad stuff). In general, it is easier to have high trust in small, relatively homogeneous groups. As teams get larger, more dispersed, and more diverse, it becomes harder to maintain a high trust environment. Ironically, the need doesn't diminish as the difficulty increases.

A high trust environment is desirable in all organizations. And, while it is as important or more important in technical organizations than in non-technical ones, it is harder to achieve. If we go back to the engineers' view of the world in the table above, we can see how even "neutral" politics poisons the high trust environment. If having a high trust environment is important to your organization, then you will need to move the bar up, and tolerate less "gray zone" behavior than you might otherwise accept.

Note: Once you have worked in a high trust environment, you will find it difficult to work in lower trust environments. All your instincts can turn out to be wrong, and a trusting person in a low trust environment is an easy target for political manipulators and can be made to suffer. On the other hand, it is crucial for people who have come from low trust environments to learn how to trust others once they are in a high trust environment; a lifetime of paranoia is difficult to overcome. Recent 謩gr豠from the old Soviet Union have told me that assimilating into the less "on guard" American culture constitutes a difficult adjustment.

Other Variants of Bad Politics

In addition to the eleven items I noted under the "Bad Politics" section above, there are a few other negative behaviors that can crop up. We collect them here so that they don't escape unnoticed.

One particularly noxious form of bad politics is failing to "sign up." In every process, there is a period of discussion, followed by a decision. At this point, "the polls are closed." Once the decision is made, everyone must commit to it. Continued whining and "politicking" for rejected solutions after the fact is bad politics. Instead of allowing the organization to move on and implement the solution, it continues to mire down the issue in debate. Worse, failure to commit tends to undermine the leadership structure by calling decisions into question over and over again. This form of bad politics must be eradicated whenever it crops up. Managers need to send a strong signal that "If you can't sign up, get off the team." If you have a healthy political process, this is a reasonable condition.

Another symptom of bad political behavior is obvious empire building. There are still many people out there who are caught up in the trappings and symbols of "success": rank in the organization, number of direct and indirect reports, size of budget, and so on. Organizations that overemphasize these statistical symbols reap what they sow; behaviors tend to follow the reward system. If, on the other hand, your organization rewards originality, contribution, "doing whatever it takes," and so on, independent of the individual's place in the hierarchy, then you are on the right track. In such organizations, the former behaviors tend to stick out like a sore thumb, and can be culturally discouraged. When individuals persist in them, they are mapping to different cultures that are more politicized, in which one's opinions have value according to one's place in the hierarchy as opposed to the opinion's intrinsic value. High trust organizations will ultimately reject such behavior, and if the behavior doesn't change, then the individuals who practice it must themselves be rejected. The sooner, the better.

Finally, there's a stylistic issue that is often characterized as political behavior, usually in a negative way. Some people turn every interaction into a negotiation; they are always "working the room." If nothing else, such individuals become tiresome, and tend to drain the energy of the group after awhile; people develop the habit of knowing that they have to "play defense" at all times with such individuals. I clearly run the risk of a libel suit if I try to give you an example here.⁸ But you all know the behavior, because there are certain individuals in every organization who are "annoying" in just this way. You tend to just not want to have to deal with them. Try not to be one, and coach others to avoid the practice.

Back to top

Summing Up

Political process is a part of human and organizational behavior. To disengage from it is to cede authority to those who choose to continue to play. It is also morally untenable to disengage out of abhorrence to "politics" and then later claim that you were disenfranchised.

As leaders, however, we do have an obligation to avoid the trap of letting politics intrude on purely technical decisions. Even at the national level, we have sometimes made the mistake of allowing political concerns to govern decisions that were untenable on technical merit. When facts are clear, we cannot ignore them, regardless of the political consequences. Try not to violate laws of physics.

When we engage in "politics," we need to encourage "good politics" as indicated above. We need to become wary when we see "neutral zone" behavior starting to occur. And we need to stamp out and crush, preferably through peer pressure, "bad politics." Basically bad politics corrupts the entire process, by admitting as legitimate those behaviors that are clearly unethical and of low integrity.

Note that even "neutral" or "gray zone" practices will undermine a high trust environment. The advantages of a high trust environment are so great that one should seriously consider whether the allowance of "gray zone" practices is worth the risk.

We need to ensure that in all political processes or negotiations, the "polls close," that decisions are made, that the team "signs up," and that we move forward as one to implement the decision, without continuing non-productive debate.

Finally, for those of you who fall back on the "all politicians are corrupt scoundrels" argument, I have only two words for you: Harry Truman. Truman was no technologist, but he was a superb politician and leader. But that is the subject of yet another article.

Back to top

Notes

¹ http://www-106.ibm.com/developerworks/rational/library/4662.html

² I must break with tradition here and explicitly credit and thank the editor, Mike Perrow, for this excellent formulation of my more direct but less clear articulation.

³ A pretty awful place in the nether regions of the castle. The word derives from the French oublier, to forget.

⁴ This is a very important concept, and we will return to it later in the article. High trust environments are valuable to all organizations, and we have found that they are especially important in technical organizations.

⁵ Let us note once again for emphasis the North American bias. Some or all of the gray zone behaviors might be considered OK, acceptable, dubious, or very bad by Europeans or Asians. This is where things get very complex. But in order to frame the discussion at all, I needed to define three categories. What I put into them reflects the North American point of view, which itself is an approximation.

⁶ We apologize also to Sun Tzu, who we could not work into this article at all.

⁷ We often find some curious and contradictory manifestations of engineers' value systems. For example, it is typical for engineers to feel uncomfortable in discussions with customers if they think you are not being "100% honest" in all your utterances. Here they apply what can sometimes be an overly rigorous standard. On the other hand, they are puzzled when their integrity seems to be impugned because they missed a schedule milestone. For them, this was just a technical problem, not an issue of a missed commitment and violation of trust. Understanding some of these characteristics is important in dealing with technical organizations, because the obvious mismatches with other parts of the organization can lead to communication problems.

⁸ But here is a sports analogy: In tennis, this behavior is akin to always playing at the net.

About the author

Comments (Undergoing maintenance)

Back to top

Trademarks  |  My developerWorks terms and conditions

Table of contents

• Context
• Definition
• Three Scenarios
• Politics Is Inevitable, But
• When Things Get Political
• Summing Up
• Notes
• About the author
• Comments

Local resources

• IBM Innovation Center - Austin, TX
• IBM Innovation Center - Chicago, IL
• IBM Innovation Center - Dallas, TX
• IBM Innovation Center - San Mateo, CA
• IBM Innovation Center - Waltham, MA
• Technical Briefings in North America

My developerWorks community

Dig deeper into Rational on developerWorks

Rate a product. Write a review.

Write a review for Rational products today!

Special offers